This study seeks to develop a prognostic system or "axis" to help evaluate and classify patients with severe functional psychiatric disorders. Such a prognostic axis would be important for clarifying the processes involved in determining the course of these disorders. It would also have the pragmatic value of helping to anticipate continuing impairment and provide a basis for evaluating treatment methods and program effectiveness. The hypothesis underlying this objective is that in severe functional psychiatric disorders predictor-outcome relationships exist that have been mistakenly considered specific to certain diagnoses or have remained entirely undetected, and that these relationships actually represent important processes that transcend individual diagnostic groups. To test this hypothesis and attempt to develop a prognostic classification axis, all first lifetime psychiatric admissions who come from a specified geographic area will be evaluated who are between the ages of 15 and 55. Evaluations of subjects and family members on admission and again at two-year follow-up will provide data on ten variables considered predictors of outcome and several measures of follow-up function. To evaluate the generality of the findings, a matched cohort of 100 outpatients will also be evaluated. The use of data collection techniques with demonstrated reliability and several methods for analyzing relations among predictors will help determine which variables singly or in combination predict outcome for functional psychiatric disorders regardless of specific diagnosis.